Exim4 + Mailman help

Sp00n · 13 Jan 2011 at 16:08

I have a debian box with exim4 and mailman running on a subdomain of my main domain, this handles the mailing lists. I noticed this morning that none of my lists were working and haven't been for a few weeks.

Now I've checked the logs for exim and I'm getting relay rejected:

2011-01-13 12:49:31 H=localhost (host.domain.net) [127.0.0.1] U=list F=<[email protected]> rejected RCPT <[email protected]>: relay not permitted

Unfortunately I don't really know anything about exim and I only really know the admin side of mailman.

Now I read the follow post:

http://forum.froxlor.org/index.php?...figuration-rejected-rcpt-relay-not-permitted/

And I spotted "hostlist relay_from_hosts" so I added 127.0.0.1 and all of a sudden it starts working again, so does this mean its now changed from a multiple config setup to a single one? In which case do I now have to reconfigure the whole damn thing again cause I'm concerned about the default settings.

I also noticed 3 other ip address in the hostlist which I don't recognise so I've removed those for now but I don't get what's happened.

Can anyone help?

Now my log files are empty :/

Sp00n · 14 Jan 2011 at 12:19

Mpemba Effect said:
The only thing I can think of is the Exim config file allows you can include other files like variables and Debian uses this to break down the original monolithic config file into tons on smaller seperate config files (very annoying imo). Exim's config file has and probably will always be a single file unless you use Debian and it's derivatives.

I appear to have both of these, the single "monolith" config file (exim4.conf.template) which it is now picking up the configuration from and where I added 127.0.0.1 and removed the 3 rogue IP's, as well as the separate config files (found within conf.d) which I assume it was using before this problem occurred.

What I don't understand is why it has changed.

Sp00n · 18 Jan 2011 at 09:34

Ok so it isn't fixed, its really weird, some of the lists work some of them do nothing.

I'm getting the following in the logs:

2011-01-18 09:24:59 1Pf7oA-0006Ht-B6 <= [email protected] H=web36802.mail.mud.yahoo.com
2011-01-18 09:25:01 1Pf7oA-0006Ht-B6 => list1 <[email protected]
2011-01-18 09:25:01 1Pf7oA-0006Ht-B6 Completed

so it see's it come in, then it tries to push it out to the list but I'm not getting any further entries as it should then email each individual user from what I can tell.

I'm beginning to get extremely annoyed with this and I might just bin the lot if I had something to switch it over too :\

Sp00n · 18 Jan 2011 at 10:35

Yea they are the same subdomain, sorry I should have written out the log file the same.

Just run some more tests following this blog: http://tperspective.blogspot.com/search/label/mailman

If I run the command:

Code:

exim -d -v [email protected]

I get a "posting held for approval" email so it looks like the list works when using that command, just not when I actually want to email anything to it.

All the other tests in that blog come back as being fine so I don't know wtf is going on

Sp00n · 19 Jan 2011 at 09:10

Ok so today I get in and everything seems fine, WTF?

Sp00n · 19 Jan 2011 at 12:19

It seems all the lists are working except a couple, does that sound plausible?

I might try deleting and re-creating them to see if that kicks them in to life.